acc=history.history['acc']
val_acc=history.history['val_acc']
loss=history.history['loss']
val_loss=history.history['val_loss']
epochs=range(len(acc)) # Get number of epochs
but, my Google Collabs doesn’t want to take “accuracy” as the key. It always getting back KeyError for accuracy. I think, this ‘acc’ and ‘accuracy’ are causing my 0 score.
Any advice to handle this? because I always getting 0/50 for my history file, and doesn’t know how to trace it
Where are you getting the KeyError? When training or when retrieving the values of the history?
If it is at training time it is most likely a TF bug. Could you please check the TF version in your colab? You can do so by running this command: ! pip show tensorflow.
If it is when retrieving the values of the history make sure that accuracy is being used when compiling the model.